Lably
Skip to content

Warming stripes

Data visualization graphics depicting trends of annual temperature changes

  • 14 min read
  • 4 chapters

Related on Lably

Through Data and information visualization, Climate change and Biodiversity loss

British climatologist Ed Hawkins originated the concept of visualizing historical temperature data in 2018. The data visualization technique was soon extended to involve animation, and to visualize diverse quantities such as sea level rise predictive climate data, progression of ocean depths, aviation's greenhouse gas emissions, biodiversity loss, soil moisture changes and fine particulate matter concentrations. The graphic has been used to visually juxtapose temperature trends with other data—such as atmospheric CO2 concentration, global glacier retreat, and precipitation.

In less technical contexts, the graphic has been embraced by climate activists, used as cover images of books and magazines, used in fashion design, and projected onto natural landmarks. It has been used on athletic team uniforms, music festival stages, and public infrastructure.

Background, publication and content

In May 2016, to make visualizing climate change easier for the general public, University of Reading climate scientist Ed Hawkins created an animated spiral graphic of global temperature change as a function of time, a representation said to have gone viral. Jason Samenow wrote in The Washington Post that the spiral graph was "the most compelling global warming visualization ever made", before it was featured in the opening ceremony of the 2016 Summer Olympics.

Separately, by 10 June 2017, Ellie Highwood, also a climate scientist at the University of Reading, had completed a crocheted "global warming blanket" that was inspired by "temperature blankets" representing temperature trends in respective localities. Hawkins provided Highwood with a more user friendly colour scale to avoid the muted colour differences present in Highwood's blanket. Independently, in November 2015, University of Georgia estuarine scientist Joan Sheldon made a "globally warm scarf" having 400 blue, red and purple rows, but could not contact Hawkins until 2022. Both Highwood and Sheldon credit as their original inspirations, "sky blankets" and "sky scarves" which are based on daily sky colours.

On 22 May 2018, Hawkins published graphics constituting a chronologically ordered series of blue and red vertical stripes that he called warming stripes. Hawkins, a lead author for the IPCC 6th Assessment Report, received the Royal Society's 2018 Kavli Medal, in part "for actively communicating climate science and its various implications with broad audiences".

As described in a BBC article, in the month the big meteorological agencies release their annual climate assessments, Hawkins experimented with different ways of rendering the global data and "chanced upon the coloured stripes idea". When he tried out a banner at the Hay Festival, according to the article, Hawkins "knew he'd struck a chord". The National Centre for Atmospheric Science (UK), with which Hawkins is affiliated, states that the stripes "paint a picture of our changing climate in a compelling way. Hawkins swapped out numerical data points for colours which we intuitively react to".

Others have called Hawkins's warming stripes "climate stripes" or "climate timelines".

Warming stripe graphics are reminiscent of colour field painting, a style prominent in the mid 20th century, which strips out all distractions and uses only colour to convey meaning. A warming stripes graphic in a 2025 Museum of Modern Art exhibit has been compared to a painting by Morris Louis–one of the earliest proponents of the colour field monument. Colour field pioneer Barnett Newman said he was "creating images whose reality is self-evident", an ethos that Hawkins is said to have applied to the problem of climate change.

Collaborating with Berkeley Earth scientist Robert Rohde, on 17 June 2019 Hawkins published for public use, a large set of warming stripes on ShowYourStripes.info. Individualized warming stripe graphics were published for the globe, for most countries, as well as for certain smaller regions such as states in the US or parts of the UK, since different parts of the world are warming more quickly than others.

Data sources and data visualization

Warming stripe graphics are defined with various parameters, including:

  • source of dataset (meteorological organization)
  • geographical scope of measurement (global, country, state, etc.)
  • time period (year range, for horizontal "axis")
  • temperature range (range of anomaly (deviation) about a reference or baseline temperature)
  • colour palette (usually, shades of blue and red),
  • colour scale (assignment of colours to represent respective ranges of temperature anomaly),
  • temperature boundaries (temperature above which a stripe is red and below which is blue, usually determined by an average annual temperature over a "reference period" or "baseline" of usually 30 years).

Hawkins' original graphics use the eight most saturated blues and reds from the ColorBrewer 9-class single hue palettes, which optimize colour palettes for maps and are noted for their colourblind-friendliness. Hawkins said the specific colour choice was an aesthetic decision ("I think they look just right"), also selecting baseline periods to ensure equally dark shades of blue and red for aesthetic balance. Hawkins chose the 1971–2000 average as a boundary between reds and blues because the average global temperature in that reference period represented the mid-point in the warming to date.

A Republik analysis said that "this graphic explains everything in the blink of an eye", attributing its effect mainly to the chosen colors, which "have a magical effect on our brain, (letting) us recognize connections before we have even actively thought about them". The analysis concluded that colors other than blue and red "don't convey the same urgency as (Hawkins') original graphic, in which the colors were used in the classic way: blue=cold, red=warm."

ShowYourStripes.info cites dataset sources Berkeley Earth, NOAA, UK Met Office, MeteoSwiss, DWD (Germany), specifically explaining that the data for most countries comes from the Berkeley Earth temperature dataset, except that for the US, UK, Switzerland & Germany the data comes from respective national meteorological agencies.

For each country-level #ShowYourStripes graphic (Hawkins, June 2019), the average temperature in the 1971–2000 reference period is set as the boundary between blue (cooler) and red (warmer) colours, the colour scale varying +/- 2.6 standard deviations of the annual average temperatures between 1901 and 2000. Hawkins noted in 2019 that the graphic for the Arctic "broke the colour scale" since it is warming more than twice as fast as the global average. He later reported that the 2023 global average was so extreme that a new, darker shade of red was required.

For statistical and geographic reasons, it is expected that graphics for small areas will show more year-to-year variation than those for large regions. Year-to-year changes reflected in graphics for localities result from weather variability, whereas global warming over centuries reflects climate change.

The NOAA website warns that the graphics "shouldn't be used to compare the rate of change at one location to another", explaining that "the highest and lowest values on the colour scale may be different at different locations". Further, a certain colour in one graphic will not necessarily correspond to the same temperature in other graphics.

A climate change denier generated a warming stripes graphic that misleadingly affixed Northern Hemisphere readings over one period to global readings over another period, and omitted readings for the most recent thirteen years, with some of the data being 29-year-smoothed—to give the false impression that recent warming is routine. Calling the graphic "imposter warming stripes", meteorologist Jeff Berardelli described it in January 2020 as "a mishmash of data riddled with gaps and inconsistencies" with an apparent objective to confuse the public.

Sign in to keep reading

3 more chapters are free with a Lably account.

More in Statistics